A manufacturing workshop AGV dynamic scheduling method, system, device and storage medium based on digital twin

CN113867295BActive Publication Date: 2024-11-26XI AN JIAOTONG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111182620.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-10-11
Publication Date
2024-11-26
Estimated Expiration
2041-10-11

AI Technical Summary

Technical Problem

In the prior art, AGV will continue to operate in the workshop transportation system until the power is below the threshold and then go to charge, resulting in uncontrollable charging timing and charging time, severely delaying the scheduling plan, and difficulty in sensing the task execution status.

Method used

A digital twin-based method is adopted to establish an AGV scheduling model, including a task scheduling model and a charging scheduling model, a genetic algorithm is used to generate a scheduling scheme, and a dynamic energy consumption model and a running rule model are combined to perform simulation verification and charging process optimization.

Benefits of technology

Dynamic scheduling of AGV is realized, ensuring that the scheduling plan is more realistic, solving the charging bottleneck problem, improving the workshop logistics efficiency, and avoiding charging delays and state perception difficulties.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113867295B_ABST
    Figure CN113867295B_ABST
Patent Text Reader

Abstract

The present invention discloses a method, system, device and storage medium for dynamic scheduling of AGV in a manufacturing workshop based on digital twins, which mainly includes six steps: establishing a digital twin model of AGV in a manufacturing workshop, establishing an AGV scheduling model, generating an initial scheduling plan, simulating and verifying the scheduling plan based on the digital twin model, predicting the AGV availability based on a dynamic energy consumption model, and charging scheduling. The digital twin technology is applied to the field of AGV scheduling, and the characteristics of virtual and real data interaction in the digital twin technology are utilized to establish a data model that accurately describes the energy consumption characteristics of the AGV, providing knowledge support for the generation of scheduling plans; the dynamic scheduling based on the digital twin considers that the generation of the AGV scheduling plan is divided into two stages: task scheduling and charging scheduling, so as to ensure that the scheduling plan is more in line with reality; the digital twin technology can also simulate and verify the generated scheduling plan, so that the scheduling plan can meet the needs of the order task.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of dynamic scheduling of AGVs in manufacturing workshops, and relates to a dynamic scheduling method, system, equipment and storage medium of AGVs in manufacturing workshops based on digital twins. Background Art

[0002] AGV scheduling optimization is a key process in manufacturing systems. Its main task is to optimize the allocation and timing management of transportation tasks in order to achieve goals such as reducing completion time, reducing energy consumption, and reducing resource idle rate. However, due to the shortcomings of existing methods such as poor resource predictability and difficulty in perceiving task execution status, AGV scheduling still has an unavoidable bottleneck link-AGV charging. For example, in the existing workshop transportation system, the AGV will continue to run until it falls below the power threshold, and then go to the charging area to charge until it is full, but the timing and duration of this charging behavior are usually uncontrollable, and in most cases will seriously delay the scheduling plan. Therefore, a reasonable AGV scheduling system should not only allocate transportation tasks, but also be able to flexibly allocate the charging process of the AGV. However, flexible allocation of charging requires predicting the future energy consumption curve of AGV in advance, so as to reasonably arrange the scheduling plan of AGV. Digital twin technology can integrate and analyze a large amount of real data collected on-site and the data of virtual models, and establish a dynamic energy consumption model that can accurately describe the energy consumption characteristics of AGV to predict the energy consumption status of AGV during operation; at the same time, the digital twin technology can be used to simulate and verify the generated AGV scheduling plan, so that the scheduling plan can meet the task requirements. Summary of the invention

[0003] In order to overcome the shortcomings of the above-mentioned prior art, the purpose of the present invention is to provide a dynamic scheduling method, system, device and storage medium for AGV in a manufacturing workshop based on digital twins, aiming to solve the defective technical problem that in the workshop transportation system in the prior art, the AGV will continue to run until the power level is lower than the threshold, and then go to the charging area to charge until it is full, so that the timing and duration of the charging behavior are usually uncontrollable, which seriously delays the scheduling plan and makes it difficult to perceive the task execution status.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] The present invention proposes a method for dynamic scheduling of AGVs in a manufacturing workshop based on digital twins, comprising the following steps:

[0006] S1. Establish an AGV scheduling model, which includes a task scheduling model and a charging scheduling model;

[0007] S2, taking the given order task information as the input of the task scheduling model obtained in S1, and using the genetic algorithm to solve the task scheduling model under the premise of satisfying relevant constraints, obtaining the task allocation and the corresponding execution order, and generating an initial task scheduling plan;

[0008] S3. Determine the status of the initial task scheduling solution obtained in S2:

[0009] The initial task scheduling plan is in an available state, and the initial task scheduling plan continues to be executed;

[0010] There is an unavailable state in the process of executing the initial task scheduling plan. Based on S2, the given order task information is used as the input of the charging scheduling model established by S1. Under the premise of meeting the relevant constraints, the genetic algorithm is used to solve the charging scheduling model to obtain the charging timing and charging amount of each AGV, thereby obtaining a charging scheduling plan including the charging process.

[0011] Preferably, the specific operation steps of S1 are as follows:

[0012] S1.1. The task scheduling model and the charging scheduling model have different decision variables, which are specifically expressed as follows:

[0013] 1) Task scheduling model decision variables

[0014]

[0015]

[0016] Among them, A a represents the AGV numbered a, i represents the i-th order, O ij Indicates the j-th transport task of order i, i is the order number; j is the process number; i' and j' are data different from order numbers i and j; Indicates the distribution of transportation tasks; It indicates the timing management of tasks, that is, the order of tasks;

[0017] 2) Decision variables of charging scheduling model

[0018]

[0019]

[0020] in, Determine the amount of electricity to be charged in a single charging task;

[0021] S1.2, the task scheduling model and the charging scheduling model have the same objective function and constraints, which are specifically expressed as follows:

[0022] 1) Objective function

[0023] min f=MS+E

[0024] Among them, MS represents the completion period of the task, and E represents the total energy consumption of AGV;

[0025] 2) Constraints

[0026] Constraint 1: The start time of a transport task cannot be earlier than the end time of the previous task;

[0027]

[0028] Where m represents the number of AGVs, n represents the number of order tasks, and tSO ij O ij The start time, tCO ij O ij Shipping time, A a Execute O ij Go to O i'j' Time required to pick up the goods, tSO i'j' It means O i'j' The start time of

[0029] Constraint 2: The AGV’s loading start time must not be earlier than the processing completion time of this process;

[0030]

[0031] Among them, p i represents the number of corresponding processes for order i, tSPO ij O ij The processing start time of the corresponding process, tSO ij+1 Indicated as O ij+1 The start time of transportation, where O ij+1 O ij The next process; tPO ij+1 Indicated as O ij+1 The corresponding processing time, where O ij-1 O ij The previous process;

[0032] Constraint 3: Each transport task can only be assigned to one AGV;

[0033]

[0034] Constraint 4: The start time of the machine tool cannot be earlier than the completion time of the previous process;

[0035] tSPOij ≥tSO ij-1 +tCO ij-1 ,j=2,...,p i

[0036] Among them, tSPO ij O ij The processing start time of the corresponding process, tSO ij-1 O ij-1 The start time of transportation, where O ij-1 O ij The previous process; tCO ij-1 O ij-1 The time required for completion is: ij-1 O ij The previous process;

[0037] Constraint 5: The power of the AGV must not be lower than the minimum threshold when performing the task;

[0038]

[0039] in, A a Real-time battery power percentage, rt is a whole subscript, indicating real-time time, d + is an adjustable positive number;

[0040] Constraint 6: The AGV’s single charge capacity must not exceed the remaining charge capacity;

[0041]

[0042] Preferably, the specific operation steps of S2 are as follows:

[0043] For a given order task information, the given order task information is used as the input of the task scheduling model, and a genetic algorithm is used to solve the task scheduling model under the premise of satisfying relevant constraints;

[0044] S2.1. The solution of the task scheduling model is encoded in the form of chromosomes. The chromosomes are divided into two sections. The two sections of chromosomes jointly determine the corresponding relationship between AGV and tasks and the execution order of tasks.

[0045] S2.2. Calculate the fitness function value F of the current population, which is specifically expressed as:

[0046] F=1 / f

[0047] Among them, f is the comprehensive value of the optimization target in the scheduling model;

[0048] S2.3, select the parent individuals in the population, and select the parent individuals by roulette;

[0049] S2.4, after crossover and mutation, the parental individuals produce chromosomes that do not conform to the regulations, which are repaired according to the given task order information and AGV information to become chromosomes that conform to the regulations;

[0050] Repeat S2.2-S2.4 to obtain the optimal individual of the iterative process, and use the optimal individual as the initial task scheduling plan.

[0051] Preferably, verification of the initial task scheduling scheme:

[0052] 1) Establishment of AGV digital twin model: Use 3D modeling software to establish the virtual entity model of AGV, define the operation logic of the virtual entity model of AGV, so that the virtual entity model of AGV can simulate the scheduling process of AGV cluster in the virtual environment according to the generated scheduling plan;

[0053] According to the AGV running speed and load data collected from the physical system, a dynamic fitting method is adopted, with the AGV speed and load as input and the energy consumption rate as output, to establish the dynamic energy consumption model of AGV, which can be specifically expressed as:

[0054]

[0055]

[0056]

[0057] in, is the estimated value of the energy consumption rate, mass and v are the load and speed of the AGV respectively, β and κ are the corresponding coefficients, χ is the offset, μ is the actual energy consumption rate, SOC is the battery state of charge, and e is the estimated error of the energy consumption rate;

[0058] The dynamics of the dynamic energy consumption model are as follows: when the estimated error e of the energy consumption rate is less than the threshold, the dynamic energy consumption model remains unchanged; when the estimated error e of the energy consumption rate is greater than or equal to the threshold, the parameters β, κ, and χ of the dynamic energy consumption model will be updated;

[0059] Establish an AGV operation rule model: define the quantity of goods that the AGV can carry at one time, the AGV power threshold, and the task delivery time T. Use the quantity of goods that can be carried at one time, the power threshold, and the task delivery time as AGV operation rules to constrain the AGV's handling process;

[0060] 2) Simulation and verification of digital twin models:

[0061] For the generated initial task scheduling plan, based on the three-dimensional virtual model and the set operation logic, the entire process of the initial task scheduling plan is simulated to obtain the completion time MS of the initial task scheduling plan s ;

[0062] Determine the completion time MS s Whether the value exceeds the delivery time requirement of this batch of orders, that is, whether it meets the MS s ≤T;

[0063] If MS is met s ≤T order task requirements, the initial task scheduling plan is sent to the execution system of the workshop;

[0064] If you do not meet the MS s ≤T order task requirements, regenerate the initial task scheduling plan.

[0065] Preferably, the dynamic energy consumption model availability prediction of AGV is:

[0066] For the AGV group that is executing the initial task scheduling plan: first, update the AGV dynamic energy consumption model; second, predict the power curve of each AGV in the future operation process based on the initial task scheduling plan and the updated AGV dynamic energy consumption model; if the power is lower than 20%, the initial task scheduling plan is in an unavailable state, and the initial task scheduling plan continues to be executed; if the power is higher than 20%, the initial task scheduling plan is in an available state, thereby obtaining the future availability result of the AGV; finally, determine whether the process of the initial task scheduling plan contains an unavailable state; if the initial task scheduling plan does not contain an unavailable state, continue to execute the initial task scheduling plan; if the initial task scheduling plan contains an unavailable state, obtain a charging scheduling plan that includes a charging process.

[0067] Preferably, the steps of obtaining a charging scheduling plan including a charging process are as follows:

[0068] Charging scheduling solution: Genetic algorithm is used. In the encoding stage, the two segments of the chromosome are used to represent the execution time of the charging task and the single charging power respectively. After multiple iterations, the charging scheduling solution of AGV can be obtained.

[0069] Obtaining a charging scheduling scheme including the charging process: Based on the charging scheduling scheme, the corresponding charging process is inserted into the task scheduling scheme to obtain a charging scheduling scheme including the charging process.

[0070] Preferably, the individuals are sorted according to the values ​​of the fitness function, and the individual with the largest value is the optimal individual in the iterative process.

[0071] The present invention also proposes a system for a dynamic scheduling method of AGV in a manufacturing workshop based on digital twin, which is characterized by comprising:

[0072] Model building module, used to build task scheduling model and charging scheduling model;

[0073] The model solving module is used to take the given order task information as the input of the task scheduling model obtained by S1, and use the genetic algorithm to solve the task scheduling model under the premise of meeting the relevant constraints, obtain the task allocation and the corresponding execution order, and generate the initial task scheduling plan;

[0074] The scheduling scheme availability judgment module is used to determine whether the initial task scheduling scheme is available during the execution of the initial task scheduling scheme, and continue to execute the initial task scheduling scheme; if there is an unavailable state during the execution of the initial task scheduling scheme, the given order task information is used as the input of the charging scheduling model established by S1 on the basis of S2, and the genetic algorithm is used to solve the charging scheduling model on the premise of meeting the relevant constraints, so as to obtain the charging time and charging amount of each AGV, thereby obtaining a charging scheduling scheme including the charging process.

[0075] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of a dynamic scheduling method for AGVs in a manufacturing workshop based on digital twins when executing the computer program.

[0076] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for dynamic scheduling of AGVs in a manufacturing workshop based on digital twins.

[0077] Compared with the prior art, the present invention has the following beneficial effects:

[0078] The present invention proposes a dynamic scheduling method for AGVs in manufacturing workshops based on digital twins. First, an AGV scheduling model is established. The AGV scheduling model includes a task scheduling model and a charging scheduling model. The given order information is used as the input data of the task scheduling model. Then, the task scheduling model is solved in combination with constraints and genetic algorithms to obtain a task scheduling solution including task allocation and corresponding execution order. Secondly, the availability of the task scheduling solution is determined to meet scheduling solutions in different states. When the task scheduling solution is in an available state, the AGV can continue to execute the initial task scheduling solution. When an unavailable state exists during the execution of the initial task scheduling solution, the AGV can continue to execute the initial task scheduling solution. , taking the given order information as the input data of the charging scheduling model initially established, and then combining the relevant constraints and genetic algorithm to solve the charging scheduling model, thereby obtaining a charging scheduling scheme including the charging timing and charging amount of each AGV. The dynamic scheduling method proposed in the present invention can realize the dynamic scheduling of AGV through the above scheme, and can also ensure that the scheduling scheme is more in line with reality, solving the technical problems existing in the prior art that in the workshop transportation system, the AGV will continue to run until the power threshold is lower than the power threshold, and then go to the charging area to charge until it is full, making the timing of the charging behavior and the charging time usually uncontrollable, seriously delaying the scheduling plan, and making it difficult to perceive the task execution status.

[0079] Furthermore, based on the digital twin platform, a dynamic energy consumption model that can accurately describe the energy consumption characteristics of AGV can be established to provide knowledge support for the generation of AGV scheduling solutions;

[0080] Furthermore, the task scheduling model and the charging scheduling model have the same objective function and constraints. Both the task scheduling model and the charging scheduling model belong to the scheduling model of AGV, so they have the same objective function and constraints. Considering the two separately can more completely describe the AGV operation process; if the objective functions and constraints of the two models are different, the obtained scheduling solution cannot guarantee its quality and performance, and is not in line with reality.

[0081] Furthermore, the operation rule model is a module under the digital twin model. The operation rule model is the premise and basis for the simulation verification of the AGV scheduling plan. It is an important part of the digital twin part of this patent. Without the establishment of this model, the AGV scheduling plan cannot be verified, which is not conducive to the application of this patent.

[0082] Furthermore, the quantity of goods transported at one time, the power threshold and the delivery time of the task are used as AGV operation rules to constrain the AGV's transportation process. This part is the operation rule model, which refines the application of digital twin simulation in AGV scheduling.

[0083] Furthermore, if the constraints are not met, it cannot be called a scheduling model that accurately describes the scheduling behavior of AGVs.

[0084] Furthermore, in the scheduling model, not only transportation tasks can be assigned, but also reasonable charging tasks can be assigned, so that the obtained scheduling plan is closer to the real scenario; the scheduling plan can be simulated and verified with the help of digital twin technology before execution, so that the scheduling plan can meet the needs of order tasks.

[0085] Furthermore, accurate prediction of AGV energy consumption, reasonable allocation of the charging process, and simulation verification of the scheduling plan can solve the charging bottleneck problem of AGV in scheduling and effectively improve the logistics efficiency of the workshop.

[0086] The present invention also proposes a system for dynamic scheduling method of AGV in manufacturing workshop based on digital twin, which adopts modularization concept to realize first-arrival picking, makes each module independent of each other, and facilitates unified management of each module. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 It is a schematic flow chart of the method for dynamic scheduling of AGV in a manufacturing workshop based on digital twins of the present invention;

[0088] Figure 2 A Gantt chart of the initial scheduling solution in the verification case of the present invention;

[0089] Figure 3 This is a Gantt chart of the final scheduling solution when the traditional AGV dynamic scheduling method is adopted in the present invention;

[0090] Figure 4 This is the Gantt chart of the final scheduling solution for the AGV dynamic scheduling method based on digital twins of the present invention. DETAILED DESCRIPTION

[0091] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0092] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0093] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0094] The present invention may be implemented in many different forms and should not be considered limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. In the accompanying drawings, components are exaggerated for clarity. Figure 1 As shown, the present invention proposes a method for dynamic scheduling of AGVs in a manufacturing workshop based on digital twins, comprising the following steps:

[0095] S1. Establish an AGV scheduling model, which includes a task scheduling model and a charging scheduling model;

[0096] S2, taking the given order task information as the input of the task scheduling model obtained in S1, and using the genetic algorithm to solve the task scheduling model under the premise of satisfying relevant constraints, obtaining the task allocation and the corresponding execution order, and generating an initial task scheduling plan;

[0097] S3. Determine the status of the initial task scheduling solution obtained in S2:

[0098] The initial task scheduling plan is in an available state, and the initial task scheduling plan continues to be executed;

[0099] There is an unavailable state in the process of executing the initial task scheduling plan. Based on S2, the given order task information is used as the input of the charging scheduling model established by S1. Under the premise of meeting the relevant constraints, the genetic algorithm is used to solve the charging scheduling model to obtain the charging timing and charging amount of each AGV, thereby obtaining a charging scheduling plan including the charging process.

[0100] Preferably, S1, an AGV scheduling model is established, and the scheduling model includes a task scheduling model and a charging scheduling model. First, a task scheduling model with the shortest completion time and the minimum AGV energy consumption as the optimization objectives is established, and the correspondence between tasks and AGVs and the execution order of tasks are used as decision variables; then, a charging scheduling model with the shortest completion time and the minimum AGV energy consumption as the optimization objectives is established, and the allocation of charging tasks and the allocation of charging amounts are used as decision variables; both the task scheduling model and the charging scheduling model use the order of machine tool processing, the working power threshold of the AGV, and the maximum capacity of the AGV battery as constraints;

[0101] S2. Initial task scheduling scheme generation: for a batch of given order task information, the given order task information is used as the input of the task scheduling model. Under the premise of satisfying relevant constraints, the task scheduling model is solved by using a genetic algorithm to obtain the task allocation and the corresponding execution order, and generate an initial task scheduling scheme;

[0102] S3. Establishment of the digital twin model of AGV in the manufacturing workshop: First, a virtual entity model of the AGV is established using 3D modeling software to define its operation logic. Then, a dynamic energy consumption model that accurately describes the energy consumption characteristics of the AGV is established based on the AGV operation speed and load. Finally, an AGV operation rule model is established to define the number of goods that the AGV can carry at one time, the AGV power threshold, and the task delivery time T, which are used as AGV operation rules to constrain the AGV's handling process.

[0103] Scheduling plan simulation and verification based on the digital twin model. For the generated initial task scheduling plan, first simulate the execution process of the scheduling plan based on the 3D solid model to obtain the task completion time in the simulation environment; then evaluate the completion time obtained in the simulation environment; if the requirements of the order task are met, the initial task scheduling plan is sent to the execution system of the workshop; if the requirements of the order task are not met, a new initial task scheduling plan needs to be regenerated;

[0104] AGV availability prediction based on dynamic energy consumption model: For the AGV group that is executing the initial task scheduling plan, first update the established AGV dynamic energy consumption model; then, predict the future power curve of each AGV based on the initial task scheduling plan and the updated AGV dynamic energy consumption model; if the power is lower than 20%, the initial task scheduling plan is determined to be unavailable; if the power is higher than 20%, the initial task scheduling plan is determined to be available, thereby obtaining the future availability result of the AGV; finally, determine whether the process of executing the initial task scheduling plan of S2 contains an unavailable state; if the initial task scheduling plan contains an unavailable state, the next step needs to be executed; if the initial task scheduling plan does not contain an unavailable state, the original initial task scheduling plan is maintained;

[0105] Charging scheduling: Based on the initial task scheduling plan of S2, first, the given order task information of the batch is used as the input of the charging scheduling model. Under the premise of meeting the corresponding constraints, the genetic algorithm is used to solve the charging scheduling model to obtain the charging timing and charging amount of each AGV, thereby obtaining a charging scheduling plan including the charging process.

[0106] The present invention provides a manufacturing workshop AGV dynamic scheduling method based on digital twin, including the following specific operation steps: wherein, the manufacturing workshop AGV scheduling involves multiple machine tools, m AGVs, and n number of order tasks:

[0107] S1. Establish the AGV scheduling model. The scheduling model includes the task scheduling model and the charging scheduling model. The optimization objectives of the task scheduling model and the charging scheduling model are to minimize the completion time and minimize the AGV energy consumption. The task scheduling and charging scheduling are based on the order of machine tool processing, the working power threshold of the AGV, and the maximum capacity of the AGV battery as constraints. The corresponding relationship between tasks and AGVs in the task scheduling model and the execution order of tasks are used as decision variables; the allocation of charging tasks and the size of a single charge in the charging scheduling model are used as decision variables. The specific implementation methods are as follows:

[0108] S1.1. The task scheduling model and the charging scheduling model have different decision variables, which are specifically expressed as follows:

[0109] 1) Task scheduling model decision variables

[0110]

[0111]

[0112] Among them, A a represents the AGV numbered a, i represents the i-th order, O ij Indicates the j-th transport task of order i, i is the order number; j is the process number; i' and j' are data different from order numbers i and j; Indicates the distribution of transportation tasks; It indicates the timing management of tasks, that is, the order of tasks;

[0113] 2) Decision variables of charging scheduling model

[0114]

[0115]

[0116] in, Determine the amount of electricity to be charged in a single charging task;

[0117] S1.2, the task scheduling model and the charging scheduling model have the same objective function and constraints, which are specifically expressed as follows:

[0118] 1) Objective function

[0119] min f=MS+E

[0120] Among them, MS represents the completion period of the task, and E represents the total energy consumption of AGV.

[0121] 2) Constraints

[0122] Constraint 1: The start time of a transport task cannot be earlier than the end time of the previous task.

[0123]

[0124] Where m represents the number of AGVs, n represents the number of order tasks, and tSO ij O ij The start time, tCO ij O ij Shipping time, A a Execute O ij Go to O i'j' Time required to pick up the goods, tSO i'j' It means O i'j' The start time of

[0125] Constraint 2: The AGV's loading start time must not be earlier than the process completion time of this operation

[0126]

[0127] Among them, p i Indicates the corresponding number of processes for order i; tSO ij+1 Indicated as O ij+1 The start time of transportation, where O ij+1 O ij The next process; tPO ij+1 Indicated as O ij+1 The corresponding processing time, where O ij-1 O ij The previous process;

[0128] Constraint 3: Each transport task can only be assigned to one AGV

[0129]

[0130] Constraint 4: The start time of the machine tool cannot be earlier than the completion time of the previous process

[0131] tSPO ij ≥tSO ij-1 +tCO ij-1 ,j=2,...,p i

[0132] Among them, tSPO ij O ij The processing start time of the corresponding process, tSO ij-1 O ij-1 The start time of transportation, where O ij-1 O ij The previous process; tCO ij-1 O ij-1 The time required for completion is: ij-1 O ij The previous process;

[0133] Constraint 5: The power of the AGV must not be lower than the minimum threshold when performing a task

[0134]

[0135] in, A a Real-time battery power percentage, rt is a whole subscript, indicating real-time time, d + It is an adjustable positive number, usually around 1 to 2. This constraint is a soft constraint.

[0136] Constraint 6: The AGV’s single charge capacity must not exceed the remaining charge capacity

[0137]

[0138] S2. Initial task scheduling scheme generation: For a given order task information, the given order task information is used as the input of the task scheduling model. Under the premise of satisfying relevant constraints, a genetic algorithm is used to solve the task scheduling model. The specific implementation method is as follows:

[0139] S2.1. The solution of the task scheduling model is encoded in the form of chromosomes. The chromosomes are divided into two sections. The two sections of chromosomes jointly determine the corresponding relationship between AGV and tasks and the execution order of tasks.

[0140] S2.2. Calculate the fitness function value of the current population, which can be specifically expressed as:

[0141] F=1 / f

[0142] Among them, f is the comprehensive value of the optimization target in the scheduling model;

[0143] S2.3, select the parent individuals in the population, and select the parent individuals by roulette;

[0144] S2.4, Crossover and mutation: After crossover and mutation, the parent individuals will produce chromosomes that do not conform to the regulations, which need to be repaired according to the order information and AGV information to make them conform to the regulations;

[0145] In this way, S2.2-S2.4 are repeatedly executed to obtain the optimal individual of the iterative process, and the optimal individual is used as the initial task scheduling solution.

[0146] Verification of the initial task scheduling scheme:

[0147] 1) Establish a digital twin model of the AGV in the manufacturing workshop. The specific implementation method is as follows:

[0148] The virtual entity model of AGV is established by using 3D modeling software, and its operation logic is defined so that the scheduling process of AGV cluster can be simulated in the virtual environment according to the generated scheduling scheme. According to the AGV running speed and load data collected from the physical system, the dynamic fitting method is adopted, with the speed and load of AGV as input and the energy consumption rate as output, to establish a dynamic energy consumption model that accurately describes the energy consumption characteristics of AGV, which can be specifically expressed as:

[0149]

[0150]

[0151]

[0152] in, is the estimated value of the energy consumption rate, mass and v are the load and speed of the AGV respectively, β and κ are the corresponding coefficients, χ is the offset, μ is the actual energy consumption rate, SOC is the battery state of charge, and e is the estimated error of the energy consumption rate.

[0153] The dynamics of the dynamic energy consumption model are as follows: when the estimated error e of the energy consumption rate is less than the threshold, the dynamic energy consumption model remains unchanged; when the estimated error e of the energy consumption rate is greater than or equal to the threshold, the parameters β, κ, and χ of the dynamic energy consumption model will be updated;

[0154] Establish an AGV operation rule model: define the quantity of goods that the AGV can carry at one time, the AGV power threshold, and the task delivery time T. Use the quantity of goods that can be carried at one time, the power threshold, and the task delivery time as AGV operation rules to constrain the AGV's handling process;

[0155] 2) Scheduling plan simulation and verification of the digital twin model: The initial task scheduling plan generated by S2 is simulated and verified; if the requirements of the order task are met, the initial task scheduling plan is sent to the execution system of the workshop; otherwise, a new task scheduling plan needs to be regenerated;

[0156] Specifically: For the generated scheduling plan, based on the three-dimensional virtual model and the established rule model, the entire process of the initial task scheduling plan is simulated to obtain the completion time MS of the initial task scheduling plan s ;

[0157] Determine the completion time MS s Whether the value exceeds the delivery time requirement of this batch of orders, that is, whether it meets the MS s ≤T;

[0158] If MS is met s ≤T order task requirements, the initial task scheduling plan is sent to the execution system of the workshop; otherwise, if it does not meet the MS s ≤T order task requirements, the initial task scheduling plan needs to be regenerated.

[0159] AGV availability prediction based on dynamic energy consumption model: For the AGV group that is executing the initial task scheduling plan: first, update the AGV dynamic energy consumption model according to the above method; then, based on the initial task scheduling plan obtained by S2 and the updated AGV dynamic energy consumption model, predict the power curve of each AGV in the future operation process; if the power is lower than 20%, the initial task scheduling plan is determined to be in an unavailable state, and the initial task scheduling plan continues to be executed; if the power is higher than 20%, the initial task scheduling plan is in an available state, thereby obtaining the future availability result of the AGV; finally, determine whether the process of the initial task scheduling plan contains an unavailable state; if the initial task scheduling plan contains an unavailable state, it is necessary to trigger the following steps to obtain a charging scheduling plan that includes the charging process; if the initial task scheduling plan does not contain an unavailable state, maintain the original initial task scheduling plan.

[0160] Charging scheduling: Based on the initial task scheduling plan of S2, a scheduling plan including the charging process is obtained, and this new scheduling plan is sent to the scheduling system. The specific implementation method is as follows:

[0161] Charging scheduling solution: Genetic algorithm is also used. Different from the solution of task scheduling solution, in the encoding stage of charging scheduling solution, the two segments of chromosome are used to represent the execution time of charging task and the single charging power respectively. The other stages of genetic algorithm are basically the same as task scheduling solution. After multiple iterations, the charging scheduling solution of AGV can be obtained.

[0162] Obtain a charging scheduling plan that includes the charging process: Based on the obtained charging scheduling plan, the corresponding charging process is inserted into the original task scheduling plan to obtain a charging scheduling plan that includes the charging process. It is worth noting that in order to ensure that a feasible solution can be obtained in the charging scheduling process, the requirements for task delivery time should be moderately relaxed.

[0163] In summary, the present invention discloses a method for dynamic scheduling of AGV in a manufacturing workshop based on digital twins, which includes six steps: establishing a digital twin model of AGV in a manufacturing workshop, establishing an AGV scheduling model, generating an initial scheduling plan, simulating and verifying the scheduling plan based on the digital twin model, predicting the AGV availability based on a dynamic energy consumption model, and charging scheduling. The present invention establishes an AGV dynamic energy consumption model based on digital twin technology, which can provide knowledge support for the generation of AGV scheduling plans; establishes a scheduling model including task scheduling and charging scheduling, which can not only allocate transportation tasks but also reasonable charging tasks in the process of scheduling plan generation, so that the obtained scheduling plan is closer to the real scene; the scheduling plan is simulated and verified with the help of digital twin technology before execution, so that the scheduling plan can meet the needs of order tasks.

[0164] In order to verify the feasibility and effectiveness of this method, an order instance was solved, and the dynamic scheduling results of AGV based on digital twins were compared with those of traditional AGVs.

[0165] like Figure 2 As shown in Figure 1, this is the Gantt chart of the initial scheduling plan for this order instance. The order instance contains 9 orders and 29 transportation tasks, which are performed by 3 AGVs. The initial power of AGV No. 3 is not enough to support its completion of the task list. When the power is lower than the specified threshold, it will go to the charging area to charge until it is full. Therefore, in order to reduce the delay time, the initial scheduling plan must be adjusted. The adjusted plan using traditional AGV dynamic scheduling is as follows Figure 3 As shown in Figure 2, the scheme after dynamic scheduling adjustment based on digital twin is as follows Figure 4 shown.

[0166] like Figure 3 As shown in the figure, when AGV No. 3 goes to the charging area to charge, the order task will be rescheduled, and the remaining unexecuted order tasks will be executed by the other two AGVs. This method can reduce the task delay caused by the charging process to a certain extent, but it still cannot meet the delivery requirements of this order instance.

[0167] like Figure 4As shown, the AGV dynamic scheduling method based on digital twins triggered the charging scheduling before the AGV had a power warning, and obtained a new scheduling scheme including the charging task. In this scheduling scheme, AGV No. 3 was charged once in an idle period before the power fell below the specified threshold (the 5-1 transportation task was completed), and only part of the power was charged. The completion time of this scheduling scheme can meet the delivery requirements of this order instance. Compared with the traditional AGV dynamic scheduling, the scheduling scheme of the present invention can effectively improve the efficiency of AGV operation and is closer to the real AGV operation scenario.

[0168] The present invention also proposes a scheduling system for a dynamic scheduling method of AGV in a manufacturing workshop based on digital twin, comprising:

[0169] Model building module, used to build task scheduling model and charging scheduling model;

[0170] The model solving module is used to take the given order task information as the input of the task scheduling model obtained by S1, and use the genetic algorithm to solve the task scheduling model under the premise of meeting the relevant constraints, obtain the task allocation and the corresponding execution order, and generate the initial task scheduling plan;

[0171] The scheduling scheme availability judgment module is used to determine whether the initial task scheduling scheme is available during the execution of the initial task scheduling scheme, and continue to execute the initial task scheduling scheme; if there is an unavailable state during the execution of the initial task scheduling scheme, the given order task information is used as the input of the charging scheduling model established by S1 on the basis of S2, and the genetic algorithm is used to solve the charging scheduling model on the premise of meeting the relevant constraints, so as to obtain the charging time and charging amount of each AGV, thereby obtaining a charging scheduling scheme including the charging process.

[0172] A schematic diagram of a terminal device provided in an embodiment of the present invention. The terminal device of this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned method embodiments are implemented. Alternatively, when the processor executes the computer program, the functions of the modules / units in the above-mentioned device embodiments are implemented.

[0173] The computer program may be divided into one or more modules / units, and the one or more modules / units are stored in the memory and executed by the processor to accomplish the present invention.

[0174] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0175] The processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0176] The memory may be used to store the computer programs and / or modules, and the processor implements various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory.

[0177] If the module / unit integrated in the terminal device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0178] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless defined as herein.

[0179] The above contents are only for explaining the technical idea of ​​the present invention and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A method for dynamic scheduling of AGV in a manufacturing workshop based on digital twin, characterized in that: The steps include: S1. Establish an AGV scheduling model. The AGV scheduling model includes a task scheduling model and a charging scheduling model. The specific steps are as follows: S1.

1. The task scheduling model and the charging scheduling model have different decision variables, which are specifically expressed as follows: 1) Task scheduling model decision variables Among them, A a Indicates the AGV numbered a, O ij Indicates the j-th transport task of order i, i is the order number; j is the process number; i' represents data different from the order number, and j' represents data different from the process number; Indicates the distribution of transportation tasks; It indicates the timing management of tasks, that is, the order of tasks; 2) Decision variables of charging scheduling model in, Determine the amount of electricity to be charged in a single charging task; S1.2, the task scheduling model and the charging scheduling model have the same objective function and constraints, which are specifically expressed as follows: 1) Objective function min f=MS+E Among them, MS represents the completion period of the task, and E represents the total energy consumption of AGV; 2) Constraints Constraint 1: The start time of a transport task cannot be earlier than the end time of the previous task; Where m represents the number of AGVs, n represents the number of order tasks, and tSO ij O ij The start time, tCO ij O ij Shipping time, A a Execute O ij Go to O i'j' Time required to pick up the goods, tSO i'j' It means O i'j' The start time of Constraint 2: The AGV’s loading start time must not be earlier than the processing completion time of this process; Among them, p i Indicates the number of corresponding processes for order i; tSO ij+1 Indicated as O ij+1 The start time of transportation, where O ij+1 O ij The next process; tPO ij+1 Indicated as O ij+1 The corresponding processing time, where O ij-1 O ij The previous process; Constraint 3: Each transport task can only be assigned to one AGV; Constraint 4: The start time of the machine tool cannot be earlier than the completion time of the previous process; tSPO ij ≥tSO ij-1 +tCO ij-1 ,j=2,...,p i Among them, tSPO ij O ij The processing start time of the corresponding process, tSO ij-1 O ij-1 The start time of transportation, where O ij-1 O ij The previous process; tCO ij-1 O ij-1 the time required for completion; Constraint 5: The power of the AGV must not be lower than the minimum threshold when performing the task; in, A a Real-time battery power percentage, rt is a whole subscript, indicating real-time time, d + is an adjustable positive number; Constraint 6: The AGV’s single charge capacity must not exceed the remaining charge capacity; S2. Take the given order task information as the input of the task scheduling model obtained in S1. Under the premise of satisfying relevant constraints, use the genetic algorithm to solve the task scheduling model, obtain the task allocation and the corresponding execution order, and generate the initial task scheduling plan. The specific operation steps are as follows: For a given order task information, the given order task information is used as the input of the task scheduling model, and a genetic algorithm is used to solve the task scheduling model under the premise of satisfying relevant constraints; S2.

1. The solution of the task scheduling model is encoded in the form of chromosomes. The chromosomes are divided into two sections. The two sections of chromosomes jointly determine the corresponding relationship between AGV and tasks and the execution order of tasks. S2.

2. Calculate the fitness function value F of the current population, which is specifically expressed as: F=1 / f Among them, f is the comprehensive value of the optimization target in the scheduling model; S2.3, select the parent individuals in the population, and select the parent individuals by roulette; S2.4, after crossover and mutation, the parental individuals produce chromosomes that do not conform to the regulations, which are repaired according to the given task order information and AGV information to become chromosomes that conform to the regulations; Repeat S2.2-S2.4 to obtain the optimal individual in the iterative process, and use the optimal individual as the initial task scheduling solution; S3. Determine the status of the initial task scheduling solution obtained in S2: The initial task scheduling plan is in an available state, and the initial task scheduling plan continues to be executed; There is an unavailable state during the execution of the initial task scheduling plan. Based on S2, the given order task information is used as the input of the charging scheduling model established by S1. Under the premise of meeting the relevant constraints, the genetic algorithm is used to solve the charging scheduling model to obtain the charging time and charging amount of each AGV, so as to obtain the charging scheduling plan including the charging process; Verification of the initial task scheduling scheme: 1) Establishment of AGV digital twin model: Use 3D modeling software to establish the virtual entity model of AGV, define the operation logic of the virtual entity model of AGV, so that the virtual entity model of AGV can simulate the scheduling process of AGV cluster in the virtual environment according to the generated scheduling plan; According to the AGV running speed and load data collected from the physical system, a dynamic fitting method is adopted, with the AGV speed and load as input and the energy consumption rate as output, to establish the dynamic energy consumption model of AGV, which can be specifically expressed as: in, is the estimated value of the energy consumption rate, mass and v are the load and speed of the AGV respectively, β and κ are the corresponding coefficients, χ is the offset, μ is the actual energy consumption rate, SOC is the battery state of charge, and e is the estimated error of the energy consumption rate; The dynamics of the dynamic energy consumption model are as follows: when the estimated error e of the energy consumption rate is less than the threshold, the dynamic energy consumption model remains unchanged; when the estimated error e of the energy consumption rate is greater than or equal to the threshold, the parameters β, κ, and χ of the dynamic energy consumption model will be updated; Establish an AGV operation rule model: define the quantity of goods that the AGV can carry at one time, the AGV power threshold, and the task delivery time T. Use the quantity of goods that can be carried at one time, the power threshold, and the task delivery time as AGV operation rules to constrain the AGV's handling process; 2) Simulation and verification of digital twin models: For the generated initial task scheduling plan, based on the three-dimensional virtual model and the set operation logic, the entire process of the initial task scheduling plan is simulated to obtain the completion time MS of the initial task scheduling plan s ; Determine the completion time MS s Whether the value exceeds the delivery time requirement of this batch of orders, that is, whether it meets the MS s ≤T; If MS is met s ≤T order task requirements, the initial task scheduling plan is sent to the execution system of the workshop; If you do not meet the MS s ≤T order task requirements, regenerate the initial task scheduling plan.

2. The method for dynamic scheduling of AGVs in a manufacturing workshop based on digital twins according to claim 1 is characterized in that: Dynamic energy consumption model availability prediction of AGV: For the AGV group that is executing the initial task scheduling plan: first, update the AGV dynamic energy consumption model; second, predict the power curve of each AGV in the future operation process based on the initial task scheduling plan and the updated AGV dynamic energy consumption model; if the power is lower than 20%, the initial task scheduling plan is in an unavailable state, and the initial task scheduling plan continues to be executed; if the power is higher than 20%, the initial task scheduling plan is in an available state, thereby obtaining the future availability result of the AGV; finally, determine whether the process of the initial task scheduling plan contains an unavailable state; if the initial task scheduling plan does not contain an unavailable state, continue to execute the initial task scheduling plan; if the initial task scheduling plan contains an unavailable state, obtain a charging scheduling plan that includes a charging process.

3. The method for dynamic scheduling of AGVs in a manufacturing workshop based on digital twins according to claim 2 is characterized in that: The steps to obtain a charging schedule that includes the charging process are as follows: Charging scheduling solution: Genetic algorithm is used. In the encoding stage, the two segments of the chromosome are used to represent the execution time of the charging task and the single charging power respectively. After multiple iterations, the charging scheduling solution of AGV can be obtained. Obtaining a charging scheduling scheme including the charging process: Based on the charging scheduling scheme, the corresponding charging process is inserted into the task scheduling scheme to obtain a charging scheduling scheme including the charging process.

4. A system for the dynamic scheduling method of AGV in a manufacturing workshop based on digital twins according to any one of claims 1 to 3, characterized in that: include: Model building module, used to build task scheduling model and charging scheduling model; The model solving module is used to take the given order task information as the input of the task scheduling model obtained by S1, and use the genetic algorithm to solve the task scheduling model under the premise of meeting the relevant constraints, obtain the task allocation and the corresponding execution order, and generate the initial task scheduling plan; The scheduling scheme availability judgment module is used to determine whether the initial task scheduling scheme is available during the execution of the initial task scheduling scheme, and continue to execute the initial task scheduling scheme; if there is an unavailable state during the execution of the initial task scheduling scheme, the given order task information is used as the input of the charging scheduling model established by S1 on the basis of S2, and the genetic algorithm is used to solve the charging scheduling model on the premise of meeting the relevant constraints, so as to obtain the charging time and charging amount of each AGV, thereby obtaining a charging scheduling scheme including the charging process.

5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the dynamic scheduling method of AGV in a manufacturing workshop based on digital twins as described in any one of claims 1 to 3 are implemented.

6. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the dynamic scheduling method of AGV in a manufacturing workshop based on digital twins as described in any one of claims 1 to 3 are implemented.